Windows Registry

[Sceptical]

Another 4

[attachment deleted by admin]

[Sceptical]

Last one. 

[attachment deleted by admin]

[Broni]

I looked at last five errors.
Two of them list NVENETFD.sys file, which is NVIDIA nForce Networking Controller Driver
Another one at nv4_mini.sys, which is NVIDIA Miniport Driver
Two others point to undescribed hardware error, and another one is about memory_corruption.

I'd start with reinstalling nVidia chipset driver.

[Sceptical]

Ok, I can try that but we did that last time as some of the original BSOD's pointed to the graphics driver.
Is it possible that the card itself is the culprit?
Could it be the motherboard?
Could it be the HDD giving up?
Power supply?
Or all or any of the above?
Cheers.

[Broni]

You can't exclude anything, but you have to start somewhere, and the best way is to deal with one possible fix at a time.

[westom]

Had bits and bobs go wrong, now it's going crazy.
I have had the sound card (on board) stop working three times.

A few others (ie Patio) have accurately identified your problem - memory or connected items that can corrupt memory.  You must execute diagnostics to 'see' the problem.  Your conclusions (memory tester reports no problem) are incomplete because you don't have the background that will explain how to find the defect.

  Defective memory can work find in a 70 degree room.  But fail intermittently with programs.  To find defective memory, a test program must also execute when memory is uncomfortable to touch - a temperature that all memory calls ideal.  IOW heat memory with a hair dryer on its highest settings.   Or in a room that is 100 degree F - a perfectly acceptable temperature to any PC.  All memory loves that temperature.  Execute the test.  Failure may be in the memory (not individual locations), in driver interface circuits on the motherboard, or in power to that memory.  But that is where you problem (with highest certainty) exists.

  Now, what else makes memory fail in strange (intermittent) ways?  A completely defective power supply ‘system’ that always boots the computer – but is defective.  Only way to eliminate those many component (in a power ‘system’) from the suspect list is a multimeter.  It must measure voltages on any purple, orange, red, and yellow wires when system is multitasking to as many peripherals (sound card, disk drive, internet, etc) as possible.  Even swapping a power supply does not identify a defect in other power ‘system’ components.

  1) Measure those voltages with a 3.5 digit multimeter.  2) Execute a good memory diagnostic (which one are you using) in the only condition that finds memory failures even before those failures cause program crashes later: heated.  Your every number (and numbers are always the most important fact) says a memory problem exists.  That does not mean just a memory SIMM.  But you have a memory problem.  Stop looking anywhere else.  You have a memory problem.

[Sceptical]

Thanks for the reply westom.
A memory problem huh? The wife is right!! 

Now then, I have read your reply a few times, if I have understood it right you are saying that the 'chain' from the power supply right to the sockets on the motherboard is where the problem lies.
So it COULD be the power supply, it COULD be the motherboard, it COULD be the RAM itself?

You have suggested two steps, step 1, ooohh, never done that before but hey, i'll give it a go, is there a guide somewhere on how to do it, (I have access to a multi meter unless a '3.5 digit' one is special?)

Step 2, sounds a bit easierI used MemTest as suggested by Patio in post 3, would this do the job or do I need a different one with the memory heated as you suggest, if I need to use the hairdryer method does this not overheat other components or should they all be happy cooking away?

Two of those BSOD's happened while attempting to start a demanding (for this computer) game so that points to what you are saying, very intermitent though, *censored* things, it's easier if it goes bang, at least you know what's wrong!!

Cheers.

[westom]

Thanks for the reply westom.
A memory problem huh? The wife is right!!  ...
You have suggested two steps, step 1, ooohh, never done that before but hey, i'll give it a go, ...
Step 2, sounds a bit easierI used MemTest as suggested by Patio in post 3, would this do the job or do I need a different one with the memory heated as you suggest,

  Step one (multimeter) is the simplest solution. Takes maybe 30 seconds.  3.5 digits means your measurements will be in three significant digits.  There is no guide since this is about as complex as removing screws to open the computer's cover.  This site may provide more information on which wires to touch and measure:
    http://www.hardwarebook.net/connector/power/atxpower.html

Step two: Memtst86 is a good memory tester.  It will test motherboard memory where your symptoms suggest failures are occurring.  It will not test video memory that can  create other (different) memory problems.

  Heat is a diagnostic tool.  Your symptoms suggest failure is not in specific memory locations; is in the interface between all memory locations and the CPU.  Heat can be used to make the intermittent failure hard.  Then be used to selectively heat items to discover where that problem exists.  Problem could be inside a memory SIMM or it could be the interface logic between SIMM and CPU.

  Suggested is to first make the problem hard.  Heat is a typical tool used to discover this.  (Excessive cooling – freezer - is another but less helpful and more difficult tool to accomplish same).

[patio]

MemTest does check video memory ....just so you know.

[Sceptical]

There is no guide since this is about as complex as removing screws to open the computer's cover.

That's only true if you have done it before, everything is easy to those that know how to do it.
I have never used a multi meter, wouldn't know what settings to use or at which end to touch which wires and then you need to know what the results mean.
I'm not knocking you, I appreciate the help but I wouldn't equate it to undoing a couple of screws.
I know the inside of a computer is a tad delicate and touching the wrong things at the wrong voltage or whatever may create a small plume of smoke!!   

[westom]

MemTest does check video memory ....just so you know.

The video memory that is 'shared' from  the motherboard?  Or video memory located on the video controller?    When was that Memtst featured added?

[westom]

That's only true if you have done it before, everything is easy to those that know how to do it.
I have never used a multi meter,

I was not kidding when I said the meter is sold where hammers are sold because it is that complex.  In reality, it takes practice to learn how to strike a hammer on a nail.  Meter is even easier.

  Set meter to voltage range - ie 20 VDC.  Touch leads to any two wires.  Read numbers on meter.  It is that complex.  Which is why meters are only sold where geniuses shop - such as K-mart.

  If it was difficult for a junior high school student, I would not have posted it.  Why did I not detail the complex facts obtained from those meter readings?  You post the numbers.  I do any complex stuff.

  If you fear a meter, then you fear jump starting a car or replacing a fluorescent light bulb.  Both are far most dangerous.  Not an exaggeration.  Actually, that risk is understated.  Jump starting, et al is far more dangerous than what I have posted.  Meter is that much easier.

  Smart is to confirm no danger exists.  Dumb is to fear only due to ignorance and blind fear.  Use the meter.  If you need more, then read the instructions – written so that any K-mart or Radio Shack shopper can understand.

[patio]

The video memory that is 'shared' from  the motherboard?  Or video memory located on the video controller?    When was that Memtst featured added?

Actually it's been there for awhle so i din't have a qualitive answer for you....
But most people miss the reported #'s anyways...